Re: Validating a program

[snow <snow@smoke.suba.com>]

>> Dale Thorn wrote:
>>|Adam Shostack wrote:
>>|put the sender in any great danger, but when the application is really serious, as it
>>|always is sooner or later, you must realize that people could be taking great risks
>>|with PGP encryption, and "pretty sure" isn't good enough when it's really, really
>>| vital to have bulletproof security.

     If it is vital to have bulletproof security, then they will:
      1) learn Cryptography and C well enough to read the code themselves.
      2) hire an expert to do 1). 
      3) Do the research and purchase a commercial package that has 
         guarentees and recommendations. 

>>        You're wrong.
>>        People can make their own choices about what level of risk
>>they're willing to accept.  That they make bad choices is not my
>>problem, except when they're paying for my opinion.
>It's easy to say, but when the "shit comes down" as they say, the average user is
>going to swear they had assurance PGP was absolutely secure, etc....

     If you believe that _anything_ is absolutely secure, you get what you 
diserve. It would seem far far cheaper to simply insert a couple extra chips
in the form of a tap in your keyboard to trap all of your keystrokes & forward
them via radio signals, or to rubber hose you. 

     PGP has been looked over by lots of people, so I trust it not to have 
any deliberate holes. As to bugs, or accidental errors, well, it is "freeware,
you get what you pay for. Sometimes you get more, and I am not denegrating 
PGP, but if you don't pay for it you shouldn't even expect it to keep working,
much less be bug free. This comes from someone whose main computer rarely 
runs commercial software (hey, free games just aren't as cool as the commercial
ones). 


Petro, Christopher C.
petro@suba.com <prefered for any non-list stuff>
snow@smoke.suba.com